期刊文献+

面向大规模数据中心的常量度数互连网络研究 被引量:4

Constant Degree Network for Massively Data Center
下载PDF
导出
摘要 如何高效互连大规模服务器是数据中心网络面临的一个重要挑战.目前提出的新型数据中心网络结构主要是通过增加服务器的网络端口数来扩展数据中心的规模,导致扩展的局限性和管理的复杂性.为此,如何设计由固定网络端口数的服务器互连而成的、具有常量度数的数据中心网络结构意义重大.提出了一种新型的面向大规模数据中心的常量度数互连网络结构CH(conjugate hypercube),该结构以固定网络端口数的服务器为中心,采用多层次互连实现了可扩展性和性能之间的平衡.理论分析和实验结果表明,该互连网络在不增加服务器网络端口数的前提下,可有效支持大规模数据中心高带宽、高容错的多模式数据通信;同时,具有良好的可部署性和可维护性. An important challenge on designing data center networking (DCN) is how to efficiently interconnect a large number of servers. Traditional tree-based structures are increasingly difficult to meet the design goals of data centers. Recently, a number of novel DCNs are proposed. However, these DCNs expand the scale of data center mainly by increasing the number of server's network interface card (NIC) ports, which brings expanding limitation and managing complexity. Consequently, it is meaningful and challenging to design a scalable structure for data centers, using only the commodity servers with fixed number of NIC ports and low-end, multi-port commodity switches. To address this problem, this paper proposes a novel DCN structure with constant degree called CH, which utilizes fixed number of NIC ports and commodity switches to interconnect large population of servers. The structure is server-centric, and leverages the expansibility and performance using multi-level interconnection. They own two potential benefits, i. e. , the expansibility and equal degree. Theoretical analysis and experiment results show that CH has excellent topology properties and can provide large data center with multi-pattern data traffic with high bandwidth and high fault- tolerance, but without increasing the number of NIC ports. Moreover, the modularity makes the structure have good deployablility and maintainability.
出处 《计算机研究与发展》 EI CSCD 北大核心 2014年第11期2437-2447,共11页 Journal of Computer Research and Development
基金 国家"八六三"高技术研究发展计划重点基金项目(2009AA012201) 国家"八六三"高技术研究发展计划基金项目(2011AA01A203)
关键词 数据中心网络 拓扑结构 常量度数 以服务器为中心 多路径 data center networking (DCN) topology structure constant degree server-centric multi-path
  • 相关文献

参考文献17

  • 1Miller C. Will big data clog networks with big traffic [EB/OL]. ( 2011- 12-13 ) [ 2012-10-09 ]. http://www. datacent erknowledge, com/arhives/2011 / 12/13/will-big-data- clog networks-with big traffic.
  • 2Hoff T. Changing architectures new datacenter networks will set your code and data free [EB/OL]. (2012-09-04) [2012- 10-09]. http://www, high scala|)ility, com/blog/2012/ 9/4/changing architectures-new-datacenter networks-will-set- your. hmal.
  • 3Manyika J, Chui M, Brown B, et al. Big data: The next frontier for innovation, competition, and productivity [R]. New York: Mekinsey Global Institute, 2011.
  • 4Kim C H. Scalable and efficient self configuring networks [D]. Princeton, N J: Princeton University, 2009.
  • 5Kliazovieh D, Bouvry P, Audzevich Y, et al. GreenCloud: A packet-level simulator of energy-aware cloud computing data centers [C] //Proc of IEEE Globecom 2010. Piscataway, NJ : IEEE, 2010:1-21.
  • 6Cisco System Incorporation. Data center: Load balancing data center services SRND, 956980 [R]. San Jose, USA: Cisco System Incorporation, 2004.
  • 7AI Fares M, Loukissas A, Vahdat A. A scalable commodity data center network architecture [C] //Proc of Special Interest Group on Data Communication. New York: ACM, 2008:63-74.
  • 8Greenberg A, Hamilton J R, Jain N, et aI. VL2: A scalable and flexible data center network [J]. Communications of the ACM, 2011, 54(3), 95-104.
  • 9Niranjan MR, Pamboris A, Farrington N, et al. PortLand: A scalable fault-tolerant layer 2 data center network fabric [C] //ProcofACMSIGCOMM'09. New York: ACM, 2009: 39-50.
  • 10Juniper Networks Incorporation. Network fabrics for the modern data center white paper. 2000327-004-EN [-R]. Sunnyvale, CA: Juniper Networks Incorporation, 2012.

二级参考文献12

  • 1http://zh.wikipedia.org/zh-cn/%Er%AT%91%E8%B5%ABNE6%9B%B2%ET%BA%BF.
  • 2Albert Greenber, Parantap Lahiri, David A Maltz, Parveen Patel, Sudipta Sengupta. Towards a next generation data archltecture: Scalability and commoditization//Proceedings of the ACM Workshop on Programmable Routers for Extensible Services of Tomorrow. Seattle, WA, USA, 2008: 57-62.
  • 3牛宪龙,陈华平,周文煜,武斌,刘晓茜.新型数据中心网络结构研究进展综述.中国国家网络年会论文集,北京,2010:35-39.
  • 4Leiserson C E. Fat- trees: Universal neworks for hardware-efficient supercomputing. IEEE Transactions on Computers, 1985, C34(10): 892-901.
  • 5Guo Chuan Xiong. DCell: A scalable and fault tolerant net work structure for data centers//Proceedings of the SIGCOMM 2008. Seattle, WA, USA, 2008:75- 86.
  • 6Guo Chuan-Xiong. BCube: A high performance, server-centric network architecture for modular data center//Proceedings of the SIGCOMM 2009. Barcelona, Spain, 2009:63- 74.
  • 7Albert Greenberg. VL2:A scalable and flexible data center network//Proceedings of the SIGCOMM 2008. Seattle, WA, USA, 2008: 51-62.
  • 8Greenberg A, Hamilton J, Maltz D A, Patel P. Cost of cloud. ACM SIGCOMM Computer Communication Review, 2009, 39(1).
  • 9Naous J, Gibb G, Bolouki S, McKeown N. NetFPGA: Reusable router architecture for experimental research//Proceedings of the PRESTO'08, 2008.
  • 10Hamilton J. Cooperative expandable micro slice servers (CEMS)//Proceedings of the 4th CIDR. Asilomar, CA, USA, 2009.

共引文献29

同被引文献38

  • 1Scott S, Abts D, Kim J, et al. The blackwidow high-radix dos network EC3 //Proc of the 33rd Annual Int Syrup on Computer Architecture. Piscataway, NJ: IEEE, 2006: 16- 28.
  • 2Faanes G, Bataineh A, Roweth D, et al. Cray cascade: A sealable HPC system based on a dragonfly network [C] // Proc of the 2012 Int Conf for High Performance Computing, Networking, Storage and Analysis. Piscataway, NJ: IEEE, 2012:1-9.
  • 3Rajamony R, Arimilli L, Gildea K. PERCS: The IBM POWER7 IH high-performance computing system [J]. IBM Journal of Research and Development, 2011, 55(3) ~ 1-12.
  • 4Binkert N, Davis A, Jouppi N, et al. The role of optics in future high radix switch design EC3//Proc of the 38th Annual Int Syrup on Computer Architecture. Piscataway, NJ: IEEE, 2011:437-447.
  • 5Kim J, Dally W, Scott S, et al. Technology-driven, highly- scalable dragonfly topology [C]//Proc of the 35th Annual Int Symp on Computer Architecture. Piseataway, NJ: IEEE, 2008: 77-88.
  • 6John S, Sudip D, John M. Exascale computing technology challenges [C] //Proc of the 9th Annual Int High Performance Computing for Computational Science. Berlin: Springer, 2011:1-25.
  • 7Kim J, Dally W, Abts D. Flattened butterfly: A cost- efficient topology for high-radix networks [C] //Proc of the 34th Annual Int Symp on Computer Architecture: New York: ACM, 2007~ 126-137.
  • 8Leiserson C. Fat-trees: Universal networks for hardware efficient supercomputing [J]. IEEE Trans on Computers, 1985, 34(10): 892-901.
  • 9Singla A, Hong C, Popa L, et al. Jellyfish: Networking data centers randomly [C] //Proc of the 9th Syrup on Networked Systems Design and Implementation. Berkeley, CA: USENIX Association, 2012: 225-238.
  • 10Shin J, Wong B, Sirer E. Small-world datacenters [C] //Proc of the 2nd Symp on Cloud Computing. New York: ACM, 2011:2:1-2:13.

引证文献4

二级引证文献12

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部